Poker-playing robots and more

steve smyth — Fri, 11 Jul 2008 23:10:17 GMT

who were the poker pros?
it makes a big difference...some are math wizards, some play the guy across from them...who lost?
if it's math guys, it may mean something...if
'table players', it don't mean much except luck...

Poker-playing robots and more

Dennis McClain-Furmansi, Dalworthington Gardens, Texas — Sat, 12 Jul 2008 18:45:28 GMT

The survey in SCIENCE says that more scientists are satisfied with their latest media coverage (though 57% is not a stellar performance). Also, there is more coverage of science in the popular media than in the past. However, the general population is no more science literate -- of either the content or the process -- than it has been in the past, and in fact seems to be losing ground. I credit all these to the same cause.

Many science articles in the popular media are written like, if not written from, press releases. They frequently present a single result as though it were a revelation or a discovery. This makes for good "headliner" copy and certainly makes the scientist(s) look good which pleases them. However, this is done as formula, without regard to the facts that the work is almost invariably the present end point of a long line of work, and may in fact be only a reaffirmation, refinement, or minor and obvious extension of previously known facts. This is hardly journalism, and I hesitate to even call it science writing, since the same process is used to cover myriad topics by writers with no knowledge of the subject beyond the fact that their editor(s) expect them to perform. This lack of knowledge translates to the average reader who is confronted with just another sensationalization and who passes through it with no more absorption than similar writings from other areas afford. I believe most scientists recognize this and are not pleased, but the recognition gained translates to greater acceptance and therefore greater grant awards, and THAT speaks the language of scientists these days. I've not just watched this happen from the outside, either.

There is still much understandable hesitation to being covered, quoted or even mentioned in the popular media, I believe for the reasons above. I was immensely tickled when the project I did my dissertation for was mentioned in "Thank You For Smoking" ("Why, they've just found that smoking can offset Parkinson's disease.") However, when I brought this up to one of the project leaders, he was fairly displeased. He neither needed the recognition nor would be much damaged by negative opinions of other researchers, sitting in an endowed chair and having been his home state's Professor Of The Year not long ago. I've since come to understand his wordless scowling response to the news much better. People sit up and notice when I mention the movie reference. Bit that's what they notice. They have no desire to understand the long lines of research that converged in this project or even the implications of the results. As a scientist, I can get more money by making use of this, but as an educator, which is what I believe every scientist is either indirectly or directly, I get nothing useful from it.

Present company is, of course, excluded. Alan's writing is true science journalism, both educational and entertaining in content. I can only hope he finds more time in his busy schedule to serve more often as an educator for future science journalists needed to counter the surge of press release (re)writers with no knowledge of the subject or of actual creative and educational writing.

Poker-playing robots and more

Dr. Darse Billings — Sat, 12 Jul 2008 22:14:31 GMT

The human players were professionals who have each earned more than a million dollars playing poker. Among them was probably the single best human player in the world in heads-up Limit Hold'em -- Matt "Hoss_TBF" Hawrilenko. Polaris took on some of the best, and won by a significant margin in overall play.

The statement "it's almost a given that some of those bots have been employed by online gaming sites" is both FALSE and irresponsible. It is in the best interest of online sites to maintain the fairest games possible, and at least two sites (PokerStars and Full Tilt Poker) spend millions of dollars to ensure the highest level of integrity in their games.

There is no doubt that some bots have been used to play poker online, but it is done by individuals, in violation of the site's terms and conditions. Polaris is the product of scientific research, and has never been used for this purpose.

- Dr. Darse Billings (founder and lead architect of the U of A Computer Poker Research Group).

Poker-playing robots and more

Alan Boyle — Sun, 13 Jul 2008 20:54:09 GMT

Thanks, Dr. Billings: If there seems to be any implication that Polaris has been used in this way, I want to strike that down immediately. I've also changed the wording of the item to reflect the view that although it's almost a given that bots have been used on poker sites, it's not a given that the bots are employed by the site operators themselves. That would be going too far, and I thank Darse for setting me straight.

Poker-playing robots and more

steve smyth — Sun, 13 Jul 2008 21:53:02 GMT

Hi Dr. Billings,
I don't know Hoss...but, there were six players according to Alan's article.
Lots of guys have won a million total.
I wondered if all were internet players, or table players.
Could the machine negate the skills of Chan, or the math wizardry of Ferguson, or David Grey?
The odds will pay off in favor of the math eventually...wouldn't they...that's what they show, eh?
Were there enough hands played that luck and skill were erased?
Did the machine have to show caution in fear of losing all its money at the table?
How was the betting determined?
Does the gadget bluff, or just play odds.
Was it most hands won, or most money won?
That's a pretty significant factor in the game.
Thanks!

Poker-playing robots and more

Kevin Moore, Rosamond, CA — Sun, 13 Jul 2008 23:32:43 GMT

come on. One on One, the computer got to win sooner or later. A real challenge is at a table with eight other players

Poker-playing robots and more

S.B. Stein E.B. NJ — Mon, 14 Jul 2008 18:25:45 GMT

Poker for humans (when playing against other people) has some level of psychology in it; reading body language, knowing some of the past of the person as well as what room is like. This could be the first step to having AI that can make the intution leaps that humans have. Did these machines bluff alot?

Poker-playing robots and more

Mitchell Wong — Sat, 26 Jul 2008 18:14:26 GMT

One of the ways they used to eliminate luck was that for each match there were two human players playing on two separate machines in two separate room. The hand dealt to the player on one machine would be dealt to the AI on the other machine, and vise-versa. The outcome of the match was determined by the combined score both players against the combined score of the machines. I believe each match consisted of 500 hands to try to reduce the luck some more.

Poker-playing robots and more

Mark — Sat, 01 Aug 2009 08:28:30 GMT

There is no doubt that some poker bots have been used to play poker online, but it is done by individuals. There is several posts in 2+2 regarding bots in full tilt poker and i dont think that Polaris is the product of scientific research, and has never been used for this purpose.